The objectives of this proposal are to advance understanding of the viral molecular determinants governing replication and virulence of flaviviruses in human hosts. Information from these studies may be used to design and improve live-attenuated viral vaccines for viruses of public health importance, as well as to gain insight into the mechanisms of viral pathogenesis. Flaviviruses are a diverse group of mosquito- and tick-transmitted RNA viruses which cause a range of disease syndromes in humans including hemorrhagic fever and acute encephalitis. Due to their continuing emergence and re-emergence on a global scale, there is growing demand for the availability of vaccines against those flavivirus infections which present unpredictable and serious threats to human populations. Because of the lack of any effective antiviral therapies against flaviviruses, vaccination in conjunction with mosquito control measures, remains the principal strategy for the prevention of disease. However there is a need to improve the efficacy and in some cases the safety of certain flavivirus vaccines. In this proposal molecular clones of yellow fever (YFV) and Japanese encephalitis (JEV) viruses will be used to define the molecular basis of neurovirulence in a mouse model, and to test the effects of attenuating mutations on the properties of experimental YFV and JEV vaccines. In Aim 1 the goal is to determine if the neurovirulence of YFV is mainly controlled by determinants encoded within the viral envelope (E) protein, both in the context of neuroadapted YF 17D vaccine, as well as the highly neurovirulent French neurotropic vaccine (FNV). In Aim 2, the YF 17D molecular clone will be modified by introduction of novel attenuating mutations in the E protein and their effects on neurovirulence properties and immunogenicity will be compared to 17D vaccine. In Aim 3 the molecular basis of JE virus neurovirulence will be investigated by constructing intertypic viruses between the attenuated JE SA14-14-2 vaccine currently used for human vaccination in China, and a highly neurovirulent JE-Nakayama strain, and testing their virulence properties in mice. The results of these studies are expected to yield data relevant to the use of molecular clone technology as an approach to vaccine development, and therefore have long-term implications for reducing the worldwide disease burden asssociated with flavivirus infections.